27.06.2013 Views

learning - Academic Conferences Limited

learning - Academic Conferences Limited

learning - Academic Conferences Limited

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

When Agents Make Suggestions About Readings<br />

Orlando Belo<br />

Algoritmi R&D Centre, University of Minho, Portugal<br />

obelo@di.uminho.pt<br />

Abstract: Significant efforts have been made during the last few years in the design and implementation of<br />

pedagogical agents for a wide range of application domains. One of the most common target area is the<br />

assistance to students in cases of regular subject studying, promoting means that help them to improve their<br />

performance and expertise in some specific subject areas. Frequently students ask their teachers about the<br />

“best” and more effective bibliographic resources that they could use to study and validate knowledge for some<br />

working topics. In this paper we will discuss the basic characteristics of pedagogical agents, approaching their<br />

typical functional architecture, and services, reinforcing the discussion on a specific class of pedagogical agents<br />

that are responsible to support students during their studying sessions, helping them in the validation of their<br />

knowledge, suggesting bibliographic resources information whenever requested.<br />

Keywords: eLearning platforms, agent based computing, intelligent tutoring systems, software agents, artificial<br />

intelligent tutors, bibliographic resources suggestion<br />

1. Introduction<br />

Agent based applications are very appellative. Software agents have been used to support a lot of<br />

tasks in real world applications (Van der Hoek & Wooldridge 2008). Ranging from telecommunications<br />

to retail, or doing monitoring services on hydroelectric power plants, agent based computing has been<br />

always a very good asset in a lot of problem solving arenas. eLearning is no exception to this<br />

attractive paradigm and to all of its characteristics and potentialities (Agarwal et al. 2004) (Leung & Li<br />

2001). One of the most relevant agent applications on this field was in the design and development of<br />

pedagogical agents, normally designed by artificial tutors or intelligent assistants. Basically, these<br />

entities are conceived to ensure more effective tutoring services, in some very specialized areas of<br />

studying, having clearly pedagogical purposes, giving assistance to students in cases of regular<br />

subject studying or even doing evaluation tasks. Additionally, they are also able to perform<br />

administrative and optimization services inside eLearning platforms, supervising what users are doing<br />

and suggesting better ways to do it or recommending particular resources that can help them in<br />

current tasks.<br />

The use of software agents as sophisticated autonomous means helping students on bibliographic<br />

resource selection seems to be very useful and appellative. The ability that agents have to adapt to<br />

new scenarios and to communicate with other means of <strong>learning</strong> makes possible a very dynamic<br />

eLearning environment, where students needs can be satisfied easily. An agent can also adapt in real<br />

time different plans of readings for current user needs, finding the best solution for a given studying<br />

plan, and personalizing processes and exploitation scenarios (Schiaffino et al. 2008). To do that, they<br />

simply need to act as search engines over their databases looking for a list of references that satisfies<br />

student current, his preferences, and a previous plan of readings prepared by his teachers. This<br />

means that any assistant agent (Okamoto et al. 2009) must establish usage profiles and accordingly<br />

prepare its plan of action an user satisfaction. It is very desirable that the processes of readings<br />

suggestion is versatile and proactive, providing the references that students require and, at the same<br />

time, giving viable alternatives that follow other suggestion indicators (e.g. readings usage ranking,<br />

recommended references, teacher’s preferences, or external sources identification).<br />

In this paper we will focus our attention on a specific class of pedagogical agents that are responsible<br />

to support students during their studying sessions, helping them in the validation of their knowledge in<br />

a particular domain, and (as their priority goal) suggesting bibliographic resources information<br />

whenever requested or inferred as necessary during an evaluation process, accordingly current<br />

studying status of the students. We will discuss the basic characteristics of these pedagogical agents<br />

(section 2), reinforcing the discussion presenting an application domain for bibliographic resource<br />

suggestion in a conventional eLearning scenario (section 3), and present the basic characteristics and<br />

functional architecture of a specific software agent for personalised assistance in bibliographic<br />

resource suggestion (section 4). Finally, in section 5, we will present some conclusions and future<br />

work.<br />

41

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!